Kiro custom agents

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

With annotations declaring openWorldHint=true but destructiveHint=false, the description adds substantial non-obvious context: a permission gate on the three write actions, the fact that an editor would be opened for editing and is therefore unavailable, and that `agent migrate` is deliberately withheld as potentially destructive. That last note is worth flagging since it explains why an otherwise plausible action is missing. It does not describe whether writes are reversible or override existing config, which keeps it off a 5.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Is the description appropriately sized, front-loaded, and free of redundancy?

Two sentences, no filler, and the most decision-relevant fact (the KIRO_MCP_ALLOW_CONFIG_WRITES gate) is stated directly rather than buried. The heavy parenthetical nesting and backtick-quoted action names make it slower to scan than ideal, but nothing is wasted.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

An output schema exists, so return shapes need not be explained, and the description covers the permission and availability constraints well. Against a 7-parameter, 5-action tool it still leaves the agent guessing about `dry_run` and the `name`/`from` arguments, which is a meaningful omission for write actions. Adequate, but not fully sufficient to invoke the write actions confidently.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

Schema description coverage is only 43%, so the description carries real burden here and does not meet it. It ties `path` to validate, but leaves `dry_run`, `name`, `from`, and `session_id` entirely unexplained beyond their type constraints — notably `dry_run`, which is schema-undocumented and materially changes write behavior. Only one parameter's meaning is added.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Does the description clearly state what the tool does and how it differs from similar tools?

The description names the resource (Kiro custom agents) and enumerates the concrete actions available (list, validate, create, set_default, upgrade), plus what `list` actually returns (advertised modes, their source, config errors). This distinguishes it clearly from the session-oriented siblings. It falls short of 5 only because the opening is dense enough that the primary read-only use (`list`) isn't front-loaded before the parenthetical.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Does the description explain when to use this tool, when not to, or what alternatives exist?

It gives real gating guidance — the write actions require KIRO_MCP_ALLOW_CONFIG_WRITES, editing is unavailable, and `agent migrate` is withheld — which tells the agent what it cannot do and why. However, it never states when an agent should reach for this tool versus kiro_settings or kiro_mcp_config, nor which action to choose for a given goal. Usage context is implied rather than directed.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.